Agent Maturity Compass API - v1.2.0
    Preparing search index...

    Function abTest

    • Run a full A/B test comparing two agent versions across all dimensions.

      Each version provides a set of ScoreSamples (one per dimension + optional "overall"). Returns per-dimension comparisons and an overall summary.

      Parameters

      • versionA: { label: string; samples: ScoreSample[] }
      • versionB: { label: string; samples: ScoreSample[] }
      • alpha: number = 0.05

      Returns ABTestResult